New to controlnets, need info on install and use.
Normally I would put the .yaml and safetensors files in the extensions folder and then select the processor and model in auto1111. SO i guess my question is what processor should I use for this model? or are there any resources out there for using this?
Normally I would put the .yaml and safetensors files in the extensions folder and then select the processor and model in auto1111. SO i guess my question is what processor should I use for this model? or are there any resources out there for using this?
As best I can tell, you go to the the \stable-diffusion-webui\models\ControlNet folder and drop this .safetensors file in. Then, with the sd-webui-ControlNet extension in Auto1111 installed when you open up controlnet under txt2img (v.1.1.224 for me currently) you leave the preprocessor blank and choose the brightness option from the model dropdown. It will then convert the grayscale image to color.
ERROR: ControlNet cannot find model config [X:\sd\stable-diffusion-webui\extensions\sd-webui-controlnet\models\control_v1p_sd15_brightness.yaml]
Got some .yaml file there mate?
ERROR: ControlNet cannot find model config [X:\sd\stable-diffusion-webui\extensions\sd-webui-controlnet\models\control_v1p_sd15_brightness.yaml]
Got some .yaml file there mate?
Try putting the safetensors in stable-diffusion-webui\models\ControlNet
Nope, still the same.
2023-06-13 18:21:49,181 - ControlNet - INFO - Loading model: control_v1p_sd15_brightness [5f6aa6ed]
2023-06-13 18:21:49,208 - ControlNet - INFO - Loaded state_dict from [X:\sd\stable-diffusion-webui\models\ControlNet\control_v1p_sd15_brightness.safetensors]
ERROR: ControlNet cannot find model config [X:\sd\stable-diffusion-webui\models\ControlNet\control_v1p_sd15_brightness.yaml]
ERROR: ControlNet will use a WRONG config [X:\sd\stable-diffusion-webui\extensions\sd-webui-controlnet\models\cldm_v15.yaml] to load your model.
ERROR: The WRONG config may not match your model. The generated results can be bad.
ERROR: You are using a ControlNet model [control_v1p_sd15_brightness] without correct YAML config file.
ERROR: The performance of this model may be worse than your expectation.
ERROR: If this model cannot get good results, the reason is that you do not have a YAML file for the model.
Solution: Please download YAML file, or ask your model provider to provide [X:\sd\stable-diffusion-webui\models\ControlNet\control_v1p_sd15_brightness.yaml] for you to download.
Hint: You can take a look at [X:\sd\stable-diffusion-webui\extensions\sd-webui-controlnet\models] to find many existing YAML files.
I'm a bit of a newbie for controlnet. I tried to inject your model .safetensor in sd-webui-controlnet but he's not happy (RuntimeError: shape '[1280, 1280, 1, 1]' is invalid for input of size 1055935)
All the models in my directory are .PTH (with an associated .yaml)
Any tips? Thanks :)
Using a copy of "cldm_v15.yaml" renamed to "control_v1p_sd15_brightness.yaml" doesn't help sadly
I mean... It works... Kinda works without yaml (I'm using this purely to make QR codes) but seeing these errors and effects makes me wonder if having that file, or not having that error, would produce better results.
As you can see in the image it works and QR code is scannable vut it makes me wonder if the results would be any better (currently I;m having problems with light colours)
.safetensors seems ok, it was just not fully uploaded to my colab I think... :)
I mean... It works... Kinda works without yaml (I'm using this purely to make QR codes) but seeing these errors and effects makes me wonder if having that file, or not having that error, would produce better results.
As you can see in the image it works and QR code is scannable vut it makes me wonder if the results would be any better (currently I;m having problems with light colours)
Exactly my question as well. I'm wondering if the results would be better if there was a .yaml file. Can't find any for this model anywhere on the net.
So... could we please get that yaml file?
请问能不能开源一个直接推理的python程序，不想用webui
Solved. use the models here: https://huggingface.co/ioclab/ioc-controlnet/tree/main/models
Oh man, you are my life saver! God bless you!